This project will use both documentary and experimental techniques to determine how the sensory-motor coordination which guides walking, develops in the weeks just before and after the start of independent walking. It will be the first study to consider how patterns of gait and the location of gaze are integrated during walking on both smooth and irregular surfaces. We will consider gait pattern, direction of focal vision, peripheral visual cues, and vestibular cues from forward locomotion. After inspecting records of normal gait and gaze patterns for evidence of coordination, we will experimentally put visual, vestibular and proprioceptive cues in conflict to determine how coordination between the sensory systems develops. We propose that there are at least two distinct ways that an infant can achieve voluntary independent walking: 1) using a deliberate step by step mode of advancing where each step is under voluntary control and, 2) using a reflex assisted stepping mode where the initiation and termination of stepping are under voluntary control, but the stepping sequence is automatic. Depending on which of these modes of stepping is first used, independent walking may begin at quite different stages of central nervous system maturation. This could alter the way independent walking should be considered in standard infant motor development scales.